in SN2 Reactions via Dynamics Mapping Deciphering Front ...publicatio.bibl.u-szeged.hu/11911/1/szabo2017.pdf3Y [Nu=F,Cl,Br,I; Y=Cl,Br,I] reactions feature a potential energy well in

Subscriber access provided by Binghamton University | Libraries

The Journal of Physical Chemistry Letters is published by the American ChemicalSociety. 1155 Sixteenth Street N.W., Washington, DC 20036Published by American Chemical Society. Copyright © American Chemical Society.However, no copyright claim is made to original U.S. Government works, or worksproduced by employees of any Commonwealth realm Crown government in the courseof their duties.

Letter

Deciphering Front-Side Complex Formationin S

N

2 Reactions via Dynamics MappingIstvan Szabo, Balazs Olasz, and Gabor Czako

J. Phys. Chem. Lett., Just Accepted Manuscript • Publication Date (Web): 09 Jun 2017

Downloaded from http://pubs.acs.org on June 11, 2017

Just Accepted

“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are postedonline prior to technical editing, formatting for publication and author proofing. The American ChemicalSociety provides “Just Accepted” as a free service to the research community to expedite thedissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscriptsappear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have beenfully peer reviewed, but should not be considered the official version of record. They are accessible to allreaders and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offeredto authors. Therefore, the “Just Accepted” Web site may not include all articles that will be publishedin the journal. After a manuscript is technically edited and formatted, it will be removed from the “JustAccepted” Web site and published as an ASAP article. Note that technical editing may introduce minorchanges to the manuscript text and/or graphics which could affect content, and all legal disclaimersand ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errorsor consequences arising from the use of information contained in these “Just Accepted” manuscripts.

1

Deciphering Front-Side Complex Formation in SN2

Reactions via Dynamics Mapping

István Szabó*,†, Balázs Olasz‡ and Gábor Czakó*,‡

†Department of Chemistry, King’s College London, London SE1 1DB, UK

‡Department of Physical Chemistry and Materials Science, Institute of Chemistry, University of

Szeged, Rerrich Béla tér 1, Szeged H-6720, Hungary

Page 1 of 19

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

2

ABSTRACT

Due to their importance in organic chemistry the atomistic understanding of bimolecular

nucleophilic substitution (SN2) reactions shows exponentially growing interest. In this

publication the effect of front-side complex (FSC) formation is uncovered via quasi-classical

trajectory computations combined with a novel analysis method called trajectory orthogonal

projection (TOP). For both F− + CH3Y [Y=Cl,I] reactions the life time distributions of the F−--

YCH3 front-side complex revealed weakly trapped nucleophiles (F−). However, only the F− +

CH3I reaction features strongly trapped nucleophiles in the front-side region of the prereaction

well. Interestingly, both back-side and front-side attack show propensity to long-lived FSC

formation. Spatial distributions of the nucleophile demonstrate more prominent FSC formation in

case of the F− + CH3I reaction compared to F− + CH3Cl. The presence of front-side intermediates

and the broad spatial distribution in the back-side region may explain the indirect nature of the F−

+ CH3I reaction.

TOC GRAPHICS

KEYWORDS mechanism, nucleophilic substitution, ion-dipole complex, front-side

intermediate, quasi-classical, trajectory projection, life time, Nu− capture

Page 2 of 19

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

3

Ion-dipole bimolecular nucleophilic substitution (SN2) is among the most prevalent reaction

pathways in chemistry and biochemistry.1-6 However, unlike what is suggested in organic

chemistry textbooks the dynamics of these reactions are quite complex.7-10 A general overall SN2

reaction can be written as Nu− + CH3Y → CH3Nu + Y−, where Nu− and Y denote the attacking

nucleophile and the leaving group, respectively. Potential energy landscape for this family of

reactions is comprised of a central barrier which separates the pre- and postreaction wells.

According to our current atomistic understanding, reactive events usually begin with the attack

of the Nu− nucleophile on the methyl side of CH3Y forming ion-dipole (Nu−--H3CY) and/or

hydrogen-bonded (Nu−--HCH2Y) pre-reaction complexes then the system goes through the

central transition state [Nu--CH3--Y]−, where synchronously a new Nu-C bond forms and the C-

Y bond breaks, while the umbrella motion around the sp3 carbon center inverts the configuration.

At higher translational energies the endothermic H-abstraction and the front-side attack channels

open. The later pathway leads to retention of the final configuration, as well as the double-

inversion mechanism revealed by our dynamics simulations8 in gas phase and recently confirmed

also in aqueous solution.11

Recently, the F− + CH3Cl and F− + CH3I SN2 reactions were probed with combined crossed-

beam imaging and molecular dynamics simulations providing interesting clues to the overall

dynamics of SN2 reactions at an atomistic level.9 Both reactions are highly exothermic and

characterized by a similar potential energy landscape with a Cs-symmetric H-bonded complex in

the entrance channel together with the traditional close-lying ion-dipole complex of C3v

symmetry connected by a low-barrier transition state. Despite of these similarities, the dynamics

of the two reactions showed substantial differences, which indicated the influence of the leaving

group in the SN2 mechanism even in the entrance channel. In case of the F− + CH3Cl SN2

Page 3 of 19

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

4

reaction we have found that at low collision energies the indirect mechanism dominates, whereas

at higher collision energies the reaction mainly occurs via the direct rebound mechanism.9,12 In

contrast, for the F− + CH3I SN2 reaction a major contribution of the indirect mechanism to the

total cross-section was observed at all collision energies investigated supported by the product

velocity, scattering angle and product internal energy distributions.9,12,13 These unexpected

qualitative differences were putatively explained based on the differences in dipole moments

which might affect the orientation of the reactants. However, the observed changes in

mechanism were not completely clarified.9 Very recently, Xie and Hase envisioned in their

perspective article10 the determining role of the front-side complexes to suppress back-side attack

and the roundabout mechanism accordingly. Front-side complexes have been recognized for

assisting in dihalide formation in the F− + CF3Br, F− + CF3I, and Cl

− + CF3Br reactions.14-16

Detailed unimolecular dynamics with lifetime distributions were determined for the [CH3--I--

OH]− front-side complex by Hase and co-workers.17 Moreover, FSCs were proposed to explain

the indirect nature of several prototypical SN2 reactions, e.g. OH− + CH3I,

17-19 and F− + CH3I,13

but the exact causes of the observed differences in the mechanism remained unclear. Our aim is

to provide a detailed characterization of the structure and energetics of the Nu−--YCH3

[Nu=F,Cl,Br,I; Y=Cl,Br,I] front-side complex minima and to explore the fascinating mechanistic

roles of these intermediate ion-dipole complexes on the example of the prototypical F−--YCH3

[Y=Cl,I] SN2 reactions. Of particular interest is the time-scale of the trapping in the front-side

pre-reaction well.

Potential energy surfaces of the Nu− + CH3Y [Nu=F,Cl,Br,I; Y=Cl,Br,I] reactions feature a

potential energy well in the front-side region, i.e. in close proximity of the leaving group.

Relative energies of the corresponding minima strongly depend on the nucleophile and also the

Page 4 of 19

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

5

leaving group. As shown in Figure 1a, the F− + CH3I reaction has a front-side complex with a

potential energy minimum of −22.8 kcal mol−1, relative to the reactants, and the equivalent FSC

minimum for the F− + CH3Cl reaction is only −2.7 kcal mol−1, whereas the corresponding values

on our chemically accurate analytical potential energy surfaces (PESs) are −22.6 and −3.7 kcal

mol−1, respectively.13 Considering all the possible Nu−--YCH3 [Nu=F,Cl,Br,I; Y=Cl,Br,I] front-

side minima (shown in Figure S1) the [F--ICH3]− has the deepest minimum, followed by the F−--

BrCH3 and Cl−--ICH3 complexes characterized by a potential energy minimum of −10.7 and

−9.2 kcal mol−1, respectively. Note that the aforementioned OH−--ICH3 long-lived intermediate

complex is characterized by an even deeper potential minimum of −26.1 kcal mol−1, predicted by

DFT based methods.17 (The corresponding value is −24.1 kcal mol−1 at the CCSD(T)-F12b/aug-

cc-pVTZ(-PP) level of theory.) Regarding the structure of F−--YCH3 [Y=Cl,I], the F-I bond

length is 0.191 Å longer than the F-Cl distance. The Cl-C/I-C bond length in F−--YCH3 [Y=Cl,I]

is 0.006/0.077 Å shorter than the corresponding bond in CH3Cl/CH3I. According to the Natural

Bond Orbital population analysis of F−--YCH3 [Y=Cl,I] the Cl and I atoms carry a partial charge

of +0.02 and +0.25, respectively, whereas the H atoms carry +0.16 for both front-side

intermediates. In contrast to the F−--ICH3 ion-dipole complex, the negative charge is distributed

disproportionately between the F and C atoms of the F−--ClCH3 complex with −0.97 and −0.52,

respectively.

To get insight into the energetics of the front-side attachment of the F− ion to the methyl-halide

in the entrance channel, the two-dimensional (2D) interaction potential was calculated by

performing a scan of the PES with YCH3 [Y=Cl,I] fixed in its equilibrium geometry. From the

2D interaction potentials depicted in Fig. 1c-f we can distinguish two deep wells for both

reactions, corresponding to the interactions of the Nu− nucleophile with the –CH3 group and the

Page 5 of 19

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

6

halogens (Y=Cl,I). It is clearly seen, that the back and front-side complex wells are separated by

a potential-ridge, which intersects the halogen at both limiting F-C-Y-H dihedral angles (see Fig.

1b). On the –CH3 side, the bottom of the potential energy well features the collinear ion-dipole

and H-bonded pre-reaction complexes, which strongly influence the migration of the Nu−

between the BSC and FSC regions of the F−--YCH3 [Y=Cl,I] SN2 reactions.9 In the front-side

region the interaction potentials substantially deviate, due to the differences in the attraction of

the two halogens. The nature of the complexation in this well is less well understood. Based on

the shape of the interaction potential we hypothesised that the potential well in the front-side

region is suitable to capture the approaching nucleophile in close proximity of the leaving group.

As we know, the trajectories may avoid even the deep potential minimum,20 thus we cannot rely

solely on the potential energy landscapes. Therefore, we are proposing an alternative approach,

called trajectory orthogonal projection (TOP) to visualize and to quantify the nucleophilic attack

in SN2 reactions. Our methodology consists of the following simple steps: (1) quasi-classical

trajectories are run at a given collision energy covering the impact parameter (b) range from 0 to

the maximum value of b, where reactive event is likely to occur, (2) the three-dimensional

positions of the nucleophile in the entrance channel are orthogonally projected to one- or two-

dimensional subspaces defined by certain atoms of CH3Y, (3) the distribution of the resulting

positions is obtained by the standard Histogram Binning technique.21 In fact, the TOP method

aligns the reactive system to subspaces defined with the nuclei in the CH3Y polyatomic reactant

providing the spatial probability of nucleophilic attack around CH3Y in a given point of the

subspace. Spatial distributions in the subspaces like the line through the {Y,C} nuclei, the

{Y,C,H} plane or the plane perpendicular to {Y,C,H} are all very informative in terms of

mapping the effective dynamics of the system. One should emphasize that TOP is capable to

Page 6 of 19

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

7

reveal reaction channels which are unlikely to follow the minimum energy path predicted by the

potential energy landscape.

With analytical potential energy functions at hand for both F− + CH3Y [Y=Cl,I] reactions8,13 we

have a unique starting point to obtain statistically accurate spatial probability distributions. To

shed light on the effective dynamics of these reactions, the trajectory orthogonal projection

(TOP) method has been applied to the entrance channel of all the reactive SN2 trajectories by

imposing the following constraints on the internal coordinates of CH3Y: rC-I < 3.5 Å, and max(rC-

H) < 2.5 Å in order to avoid the interference with the SN2 exit-channel and the proton-abstraction

pathway. (Note that the barrier of halogen abstraction is usually much higher than the maximum

collision energy in this study.) In Fig. 2 the normalized spatial probability of the attacking Nu−

around CH3Y (Y=Cl,I) is shown at collision energy of 1 kcal mol−1. On the one hand, the

distributions corresponding to the two reactions bear similarities in the back-side region, where

both reactions show a characteristic peak which centers (red spot) on the C3 axis of CH3Y

[Y=Cl,I] at a Nu−-I distance equal to the sum of rI-C,eq and rC-Nu,eq distances in the IH3C--Nu−

prereaction ion-dipole complex. As also expected, a higher probability region emerges on the 2D

map in close proximity of the H atom as an indicator of the extensive YH2CH--Nu− H-bonded

complex formation. It is noteworthy that the Nu− probability distribution in the BSC region is

broader in case of the F− + CH3I reaction, which is consistent with the larger bmax values for this

SN2 reaction.12,13 This increased chance for energy redistribution between the inter- and

intramolecular modes of the pre-reaction complex may be a principal recipient of the more

pronounced indirect nature of the F− + CH3I SN2 reaction. On the other hand, in the front-side

region the difference in Nu− spatial distributions is striking. The F− + CH3I reaction features a

prominent peak corresponding to the position of F− in the F−--ICH3 front-side ion-dipole

Page 7 of 19

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

8

complex; however, in case of the F− + CH3Cl reaction the spatial distribution of F− is completely

isotropic in the front-side region. These findings are in line with the characteristics of the

interaction potentials. The probability distributions for both reactions are qualitatively the same

up to a collision energy of 15.9 kcal mol−1 (see Fig. S3-5). At higher collision energies (Fig. S6)

the reaction is mainly direct suppressing the orientation effects, thus only those reactants lead to

products, which start with back-side attack. Moreover, the dominant direct roundabout

mechanism hinders the formation of ion-dipole and H-bonded complexes, and results in

backward scattered products.12,13

An approach for investigating the time-scale of nucleophile capture is to consider the fraction of

the trajectory spent in the FSC region.22 Considering the shape of the interaction potential, as

well as the spatial distribution of the nucleophile, the life time of the individual capture events is

calculated from the trajectory integration times in the �Nu�Y

< 0 region, where �Nu�Y is the distance

of the trapped Nu− from the Y halogen atom after orthogonal projection to the {Y,C} line. We

should emphasize, that our definition for the front-side region is not limited to the close

proximity of the Y halogen atom, because after inspection of many trajectories we had to realize

that the trapped trajectories span a long-range region of the configurational space. Nevertheless,

the exact separation of the temporarily and strongly trapped trajectory segments is not

straightforward. Although, this difficulty may eventually be overcome by analysing the residence

time of the F− ion in the FSC region. As seen on Fig. 3a and 3b, the life time distribution of

individual front-side complexation events features two regions. Up to 4 and 2 ps for F−--ICH3

and F−--ClCH3, respectively, the short-lived transient complexes are represented by high

probability peaks, but the probability instead of dropping to zero it extends to very large life

times, denoted with tFSC,max, especially in case of the F− + CH3I reaction. These long-lived,

Page 8 of 19

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

9

strongly trapped complexes have an isotropic life time distribution up to tFSC,max (not shown on

the graphs) and show little Ecoll dependence at moderate reactant translational energies. To track

the differences in life time distributions for the two systems we performed QCT simulations with

Cl of mass 127 a.u. (corresponding to I) using the F− + CH3Cl PES.23 As predicted before,9 the

mass-scaling has only minor effects on the life time distribution and tFSC,max, thereby

underpinning the role of the interaction potential. Considering the partitioning of the total life

between short-lived and long-lived FSCs the two reactions differ dramatically. Interestingly, the

strongly trapped F− + CH3I trajectories constitute only about 1% of the FSC events; but it is truly

impressive that they accumulate ~35% of the total FSC life time at a collision energy of 1.0 kcal

mol−1 and ~55% at 15.9 kcal mol−1 considering the back-side and front-side attack trajectories

together. Since the reaction becomes more direct at higher collision energies the fractions of life

time corresponding to BSA trajectories are also increased. In case of the F− + CH3Cl reaction the

fraction of long-lived trajectories is almost negligible and the Ecoll dependence of the life time

fractions shows similar trend to the F− + CH3I system.

Representative trajectories presented in Fig. 4 hold evidence for formation of the front-side

intermediate complex on the example of the F− + CH3I reaction at Ecoll = 2 kcal mol−1. As seen

on panel a, even trajectories starting with back-side attack (red arrow) can easily get around the

polyatomic reactant, CH3I and after spending a short time in the front-side well can lead to

reactive event. Another typical trajectory is shown on panels b and c projected to the

{I,C,H(red)} plane and to the plane perpendicular to {I,C,H(red)}, respectively. Here, the F−

nucleophile approaches CH3I from the halogen side and it is immediately captured in the front-

side well for 42.1 ps. Note that this residence time is approximately 20 times larger than the

trajectory integration time of a complete rebound or stripping process and it is comparable with

Page 9 of 19

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

10

the time-scale of the indirect mechanism.8,12,13 After dissociation of the F−--ICH3 intermediate

complex, the F− ion leaves the front-side region being trapped for a short time in the back-side

region of the pre-reaction well. Once its relative orientation to the permanent dipole becomes

appropriate, the substitution event can take place by simultaneous C-F bond formation and C-I

bond rupture.

Although, front-side intermediates assume a non-reactive ion-dipole orientation, the dynamics of

SN2 reactions at low collision energies is partially controlled by the nucleophile capture in the

front-side region of the pre-reaction well. At the high collision energy regime (> 35 kcal mol−1)

the probability to find front-side complexes is lower, because the FSC formation is suppressed

leading to shorter life times. We can conclude that FSC formation is a principal component of

the indirect mechanism. It will be of great interest to investigate the effect of mode-specific

excitation on the migration of the Nu− in the pre-reaction well and quantum dynamics studies are

also highly desired to analyse the resonant states in the front-side region.24 In a wider context, the

present proof of principle application of the trajectory orthogonal projection (TOP) method is a

solid starting point to reveal reaction pathways and thoroughly understand the effective dynamics

of the fundamentally important SN2 reactions using dynamics mapping.

Page 10 of 19

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

11

Figure 1. Structure and energetics of front-side complex (FSC) minima. (a) Structural

parameters and energies relative to the F− + CH3Y(eq) [Y=Cl,I] reactant asymptote obtained at

the CCSD(T)-F12b/aug-cc-pVTZ(-PP) level of theory (b-f) Entrance channel interaction

potential energy scans of F− + CH3Y(eq) [Y=Cl,I] performed at the CCSD(T)-F12b/aug-cc-

pVDZ(-PP) level of theory; the energies are relative to the F− + CH3Y(eq) reactant asymptote.

(b) Definition of the potential energy surface scan. The structural parameters and relative

energies of Nu− + CH3Y [Nu=F,Cl,Br,I; Y=Cl,Br,I] FSC minima are given in the Supporting

Information.

Page 11 of 19

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

12

Figure 2. Spatial probabilities of the F− nucleophile around CH3Y [Y=Cl,I] on the reactant side

of the F− + CH3Y → CH3Y + F− [Y=Cl,I] substitution reactions at the collision energy of 1 kcal

mol−1 using all the trajectories, which satisfy the following conditions: rC-Y < 3.5 Å and rC-H < 2.5

Å. Normalized distributions were obtained by 1D (a,b) and 2D (c,d) trajectory orthogonal

projection (TOP) to the {C,Y} line and to the {Y,C,H(red)} plain as indicated by the structures,

respectively, combined with the standard Histogram Analysis method.21 The front-side and back-

side complex regions of the configuration space are denoted with FSC and BSC, respectively.

Further spatial probability distributions at collision energies of 4.0, 15.9, 35.3 and 50.0 kcal

mol−1 are given in the Supporting Information.

Page 12 of 19

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

13

Figure 3. (a,b) Front-side complex life time distributions for the F− + CH3Y [Y=Cl,I] SN2

reactions at different collision energies. The life time is calculated from the trajectory integration

time spent by the F− nucleophile in the front-side region defined as �F�Y< 0, where �F�

Y is the

distance of F− from the Y leaving group after TOP. (c,d) Front-side complex life time fractions

split into the individual contribution of trajectories starting with front-side attack (FSA) and

back-side attack (BSA). The FSA and BSA trajectories are distinguished based on the initial

attack angle defined as the angle between the C-Y vector and the velocity vector of CH3Y at t =

0. Furthermore, the FSA and BSA are split into the contribution of short-lived (transient) and

long-lived (strongly trapped) FSC trajectories. The latter ones are characterized by a life time

larger than 2 and 4 ps for Y=Cl and Y=I, respectively.

Page 13 of 19

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

14

Figure 4. Representative trajectories in the entrance channel of the F− + CH3I SN2 reaction at

collision energy of 2 kcal mol−1 exhibiting front-side complex (FSC) formation. The dynamics

maps were obtained by projection to the planes indicated by the structures on each panel using

the 2D TOP method (see text and also Figure 2). (a) Transient short-lived FSC trajectory starting

with back-side attack, and showing F− trapped in the FSC region of the pre-reaction well for 0.6

ps before the reactive SN2 event. (b,c) Two projections of the same trajectory starting with front-

side attack, and featuring long-lived intermediate FSC; the F− nucleophile is strongly trapped for

42.1 ps before the reactive substitution event.

Page 14 of 19

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

15

COMPUTATIONAL METHODS

All the ab initio electronic structure computations (geometry optimizations, frequency

computations, and potential energy surface scans) were carried out by the Molpro 2015.1

program,25 except the computation of the molecular orbitals and the Natural Bond Orbital (NBO)

population analysis, which were done with the Gaussian 09 program package.26

Quasi-classical trajectory (QCT) computations for the F− + CH3Y [Y=Cl,I] reactions were

performed with our in-house computer code using the recently developed analytical ab initio

potential energy surfaces.8,13 The vibrational ground state of the polyatomic reactants CH3Y

[Y=Cl,I] was prepared by normal mode sampling and their rotational temperature was set to 0 K.

The initial orientation of CH3Y [Y=Cl,I] was randomly sampled and the distance between the

centers of mass of the reactants was (x2+b2)1/2, where b is the impact parameter and x was 30/40,

20/30, 20/30, 20/20 and 20/20 bohrs for Y=Cl/I at collision energies of 1, 2, 4, 10 and 15.9 kcal

mol−1, respectively. Trajectories were also run at collision energies of 35.3 and 50 kcal mol−1 for

the F− + CH3I reaction with x set to 20 bohrs. b was scanned from 0 to bmax, which is the

maximum value of b, where any reactive event can occur, with a step size of 0.5 bohr, except at

collision energies of 35.3 and 50 kcal mol−1, where a smaller step size of 0.125 bohr was

employed in order to get improved statistics for the spatial distributions of the nucleophile. At

each b 5000 trajectories were propagated, resulting in more than 2 million trajectories in this

study. We also performed dynamics simulations for the mass-scaled reaction F− + CH3Cl’ by

setting the mass of Cl’ to 127 a.u. and using the PES of F− + CH3Cl.23 For the exothermic F− +

CH3Y → CH3Y + F− [Y=Cl,I] substitution reactions the zero-point energy (ZPE) violation is

negligible, thus the QCT product analysis considered all the reactive trajectories. The spatial

Page 15 of 19

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

16

distributions of the F− nucleophile were obtained by the trajectory orthogonal projection (TOP)

method combined with the standard Histogram Analysis procedure.

ASSOCIATED CONTENT

Supporting Information

Structures and relative energies of the Nu−--YCH3 [Nu=F,Cl,Br,I; Y=Cl,Br,I] front-side

complexes (Figure S1); Molecular orbitals and atomic charges of the F−--YCH3 [Y=Cl,I] front-

side complexes (Figure S2); Spatial probabilities of the F− nucleophile around CH3Y [Y=Cl,I] on

the reactant side of the F− + CH3Y → CH3Y + F− [Y=Cl,I] substitution reactions at different

collision energies (Figures S3-6); Classical energies of the Nu−--YCH3 [Nu=F,Cl,Br,I;

Y=Cl,Br,I] front-side complex minima (Table S1); Harmonic vibrational frequencies for front-

side complex minima (Table S2)

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected] (I.S.), [email protected] (G.C.).

Notes

The authors declare no competing financial interests.

ACKNOWLEDGMENT

I.S. was supported by the UK EPSRC Fellowship EP/N020669/1. G.C. was supported by the

Scientific Research Fund of Hungary (PD-111900) and the János Bolyai Research Scholarship of

the Hungarian Academy of Sciences. We acknowledge the National Information Infrastructure

Page 16 of 19

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

17

Development Institute for awarding us access to resource based in Hungary at Debrecen and

Szeged.

REFERENCES

1. Vollhardt, K. P. C.; Shore, N. E. Organic Chemistry: Structure and Function (W. H.

Freeman, 2005).

2. Hase, W. L. Simulations of Gas-Phase Chemical Reactions: Applications to SN2

Nucleophilic Substitution. Science 1994, 266, 998-1002.

3. Chabinyc, M. L.; Craig, S. L.; Regan, C. K.; Brauman, J. I. Gas-Phase Ionic Reactions:

Dynamics and Mechanism of Nucleophilic Displacements. Science 1998, 279, 1882-

1886.

4. Mikosch, J.; Trippel, S.; Eichhorn, C.; Otto, R.; Louderaj, U.; Zhang, J.-X.; Hase, W. L.;

Weidemüller, M.; Wester, R. Imaging Nucleophilic Substitution Dynamics. Science

2008, 319, 183-186.

5. Otto, R.; Brox, J.; Trippel, S.; Stei, M.; Best, T.; Wester, R. Single Solvent Molecules

Can Affect the Dynamics of Substitution Reactions. Nature Chem. 2012, 4, 534-538.

6. Xie, J.; Otto, R.; Mikosch, J.; Zhang, J.; Wester, R.; Hase, W. L. Identification of

Atomic-Level Mechanisms for Gas-Phase X− + CH3Y SN2 Reactions by Combined

Experiments and Simulations. Acc. Chem. Res. 2014, 47, 2960-2969.

7. Brauman, J. I. Not So Simple. Science 2008, 319, 168.

8. Szabó, I.; Czakó, G. Revealing a Double-Inversion Mechanism for the F− + CH3Cl SN2

Reaction. Nat. Commun. 2015, 6, 5972.

9. Stei, M.; Carrascosa, E.; Kainz, M. A.; Kelkar, A. H.; Meyer, J.; Szabó, I.; Czakó, G.;

Wester, R. Influence of the Leaving Group on the Dynamics of a Gas-Phase SN2

Reaction. Nature Chem. 2016, 8, 151-156.

Page 17 of 19

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

18

10. Xie, J.; Hase, W. L. Rethinking the SN2 Reaction. Science 2016, 352, 32-33.

11. Liu, P.; Wang, D.; Xu, Y. A New, Double-Inversion Mechanism of the F− + CH3Cl SN2

Reaction in Aqueous Solution. Phys. Chem. Chem. Phys. 2016, 18, 31895-31903.

12. Szabó, I.; Császár, A. G.; Czakó, G. Dynamics of the F− + CH3Cl → FCH3 + Cl− SN2

Reaction on a Chemically Accurate Potential Energy Surface. Chem. Sci. 2013, 4, 4362-

4370.

13. Olasz, B.; Szabó, I.; Czakó, G. High-Level Ab Initio Potential Energy Surface and

Dynamics of the F− + CH3I SN2 and Proton-Transfer Reactions. Chem. Sci. 2017, 8,

3164-3170.

14. Cyr, D. M.; Bishea, G. A.; Scarton, M. G.; Johnson, M. A. Observation of Charge-

Transfer Excited States in the I−•CH3I, I−•CH3Br, and I

−•CH2Br2 SN2 Reaction

Intermediates Using Photofragmentation and Photoelectron Spectroscopies. J. Chem.

Phys. 1992, 97, 5911-5914.

15. Morris, R. A.; Viggiano, A. A. Kinetics of the Reactions F− and CF3Br and CF3I as a

Function of Temperature, Kinetic Energy, Internal Temperature, and Pressure. J. Phys.

Chem. 1994, 98, 3740-3746.

16. Cyr, D. M.; Scarton, M. G.; Wiberg, K. B.; Johnson, M. A.; Nonose, S.; Hirokawa, J.;

Tanaka, H.; Kondow, T.; Morris, R. A.; Viggiano, A. A. Observation of the XY−

Abstraction Products in the Ion-Molecule Reactions of X− + RY → XY− + R: An

Alternative to the SN2 Mechanism at Suprathermal Collision Energies. J. Am. Chem. Soc.

1995, 117, 1828-1832.

17. Xie, J.; Sun, R.; Siebert, M. R.; Otto, R.; Wester, R.; Hase, W. L. Direct Dynamics

Simulations of the Product Channels and Atomistic Mechanisms for the OH− + CH3I

Reaction. Comparison with Experiment. J. Phys. Chem. A 2013, 117, 7162-7178.

Page 18 of 19

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

19

18. Xie, J.; McClellan, M.; Sun, R.; Kohale, S. C.; Govind, N.; Hase, W. L. Direct Dynamics

Simulation of Dissociation of the [CH3--I--OH]− Ion-Molecule Complex. J. Phys. Chem.

A 2015, 119, 817−825.

19. Xie, J.; Otto, R.; Wester, R.; Hase, W. L. Chemical Dynamics Simulations of the

Monohydrated OH−(H2O) + CH3I Reaction. Atomic-Level Mechanisms and Comparison

with Experiment. J. Chem. Phys. 2015, 142, 244308.

20. Sun, L.; Song, K.; Hase, W. L. A SN2 Reaction That Avoids Its Deep Potential Energy

Minimum. Science 2002, 296, 875-878.

21. Szabó, I.; Czakó, G. Mode-Specific Multi-Channel Dynamics of the F− + CHD2Cl

Reaction on a Global Ab Initio Potential Energy Surface. J. Chem. Phys. 2016, 145,

134303.

22. Su, T.; Wang, H.; Hase W. L. Trajectory Studies of SN2 Nucleophilic Substitution. 7. F−

+ CH3Cl → FCH3 + Cl−. J. Phys. Chem. A 1998, 102, 9819-9828.

23. Sun, L.; Park, K.; Song, K.; Setser, D. W.; Hase, W. L. Use of a Single Trajectory to

Study Product Energy Partitioning in Unimolecular Dissociation: Mass Effects for

Halogenated Alkanes. J. Chem. Phys. 2006, 124, 064313.

24. Wang, Y.; Song, H.; Szabó, I.; Czakó, G.; Guo, H.; Yang, M. Mode-Specific SN2

Reaction Dynamics. J. Phys. Chem. Lett. 2016, 7, 3322-3327.

25. Werner, H.-J.; Knowles, P. J.; Knizia, G.; Manby, F. R.; Schütz, M.; et al. Molpro,

version 2015.1, a package of ab initio programs; see http://www.molpro.net.

26. Frisch, M. J.; et al. Gaussian 09, revision E01; Gaussian, Inc.: Wallingford, CT, 2009.

Page 19 of 19

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960